Amal Michael

This proposal describes a the MSDS692 Data Science Practicum project aimed at analyzing the U.S. broadband divide. The project's goal is to create a data-driven model that goes beyond simple mapping. I will use publicly available data from the FCC and the U.S. Census Bureau, along with geospatial feature engineering, to identify and characterize the multidimensional nature of this issue. The Expected outcome of this Project will be a predictive tool and a set of actionable insights for policymakers. The final project will provide a valuable, data-driven resource to help address the U.S. broadband divide.

This project addresses the persistent ”Digital Divide” in the United States by shifting from a binary analysis of internet availability to a granular evaluation of Infrastructure Quality and Funding ROI. Utilizing data from the FCC (Broadband Data Collection), US Census (ACS), and USASpending, this research classifies Internet Service Providers (ISPs) into three tiers based on national subscriber market power. The methodology introduces a Weighted Internet Provider Availability Score (WAS) at the Census Block Group level to measure competitive density. By analyzing federal funding allocations from 2020–2025, the project aims to identify ”investment dead zones” where high funding has not translated into improved service tiers. The anticipated outcome is a diagnostic framework that provides data-driven policy recommendations for future broadband subsidy distribution.